33results about How to "Make up for the lack of precision" patented technology

The invention discloses a method for predicting the seepage field of a nodal region of an earth and rockfill dam and determining the infiltration side of the earth and rockfill dam. The method includes establishing a nodal-region three-dimensional geological detailed model of an unfavorable geologic body and fractures; setting up three-dimensional seepage and computing a grid model on the basis of the three-dimensional geological detailed model; simulating numerical values of the seepage field of the earth and rockfill dam on the basis of the three-dimensional geological detailed model, dispersing by the finite volume method according to the determined continuity equation, the water and steam two-phase N-S equation, the water and steam two-phase mixing speed equation and mixing density equation, the set boundary conditions and the seepage parameters of the actual project, solving with the SIMPLE algorithm, realizing numerical simulation of the seepage field of the earth and rockfill dam on the basis of the three-dimensional geological detailed model and thereby predicting the seepage field of the nodal region of the earth and rockfill dam. Geological information can be accurately reflected, and the seepage field of the nodal region of the earth and rockfill dam can be accurately simulated. Meanwhile, the seepage field is utilized to determine the infiltration side of the earth and rockfill dam by the VOF (volume of fluid) method.

The invention relates to an improved face key point detection method based on GIOU and weighted NMS. Firstly, a lightweight cascaded face target detection network is adopted to carry out detection androtation angle detection on a face target in an image, GIOU is adopted to replace IOU to serve as a target frame position precision index, and IOU loss is adopted to carry out position regression; secondly, a weighted NMS suppression method is adopted for the obtained face target, and a face target box with high confidence in the image is obtained; and finally, the face key points are detected and regressed by adopting a lightweight cascaded face key point detection network. Compared with a traditional convolutional network, the speed is improved, the real-time performance of the algorithm isenhanced, and the network detection precision is improved on the premise of not additionally increasing the operand. The problem that a traditional target detection and face key point detection method is poor in real-time performance is solved, and real-time key point detection can be carried out on face targets in multiple different rotating directions in a complex environment under the condition that the detection precision is guaranteed.

The invention relates to a preparation method of a novel biodegradable rapid light curing molding material. The preparation method of the novel biodegradable rapid light curing molding material comprises the following steps: 1) performing ring-opening reaction on caprolactone which is subjected to water removal by calcium hydroxide under the initiation of polyethylene glycol to prepare a polycaprolactone-polyethylene glycol-polycaprolactone block copolymer PCEC; 2) performing double-bond modification on the product obtained in the step 1) through acryloyl chloride to prepare a double-bond endblock polymer enePCEC; and 3) performing UV irradiation on the product obtained in the step 2), a chain extender (polyethylene glycol diacrylate PEGDA), a crosslinking agent [tetra(3-thiohydracrylic acid)pentaerythritol ester] and a photoinitiator (benzoin dimethyl ether) and performing curing molding. A preparation and curing system of a photo-curing material can be improved greatly; and non-toxic, efficient and environmentally-friendly photo-curing technology and photo-curing system are constructed by taking a biodegradable polymer (caprolactone) as a raw material and combining with click chemical reaction.

The invention discloses a pre-coding method and device. The method comprises: an approximate channel h is outputted at a terminal; according to the approximate channel h, each codebook vector in a codebook is evaluated, wherein the codebook is one known by a base station and the terminal; and on the basis of the evaluation result, the codebook vectors are fed back to the base station. The base station side receives the codebook vectors fed back by the terminal, wherein the codebook vectors are obtained by evaluation on the codebook vectors in the codebook based on the approximate channel h; an analog pre-coding matrix is formed based on the codebook vectors; and an equivalent channel is obtained according to the analog pre-coding matrix. Therefore, the terminal can adjust the number of the fed-back codebook vectors dynamically to realize adaption to base stations with different digital channel numbers, so that all digital channel resources can be utilized fully. Moreover, analog and digital pre-coding precision during the analog-digital two-stage pre-coding process can be improved and the channel feedback costs are reduced.

The invention relates to an ultrasonic-assisted unmanned aerial vehicle autonomous stable landing system and method, the landing system comprises an unmanned aerial vehicle and a communication base station, and a flight controller, an ultrasonic ranging device, an inertial measurement device and an RTK positioning device are carried on the unmanned aerial vehicle; the landing method comprises thefollowing steps that the unmanned aerial vehicle flies to a preset position over a landing point; the unmanned aerial vehicle descends downwards to the preset maximum ground clearance; the ultrasonicranging device measures the ground clearance of the unmanned aerial vehicle and adjusts the landing speed of the unmanned aerial vehicle in real time according to the ground clearance until the unmanned aerial vehicle touches the ground. Compared with the prior art, the system and method have the advantages that positioning marks do not need to be arranged at landing points, accurate positioning in the horizontal direction is achieved in an RTK positioning mode, accurate positioning in the vertical direction is achieved through an ultrasonic ranging device, and accurate landing of the unmannedaerial vehicle is achieved; and the ground clearance of the unmanned aerial vehicle is measured in real time through the ultrasonic ranging device, the landing speed is adjusted in real time accordingly, and accurate control over the landing speed of the unmanned aerial vehicle is achieved.

The invention relates to a tracking system and a method for improving satellite pseudo range precision. The system comprises a frequency mixer, a carrier NCO, a code NCO, a C code generator, a first shift register, a switching circuit, a P code generator, a second shift register, an I branch correlator, a Q branch correlator, an integral sum reset module and a processor, wherein the integral sum reset module is respectively connected with the processor, the I branch correlator and the Q branch correlator; and through the switching circuit, switching between the C code and the P code is realized. According to the tracking system and the method for improving satellite pseudo range precision by adopting the structure, C code and P code signal tracking can be carried out at the same time, and after the C code is captured and tracked, the P code is controlled to start according to time information, and P code tracking is switched. Not only the C code capturing speed but also the high P code pseudo range precision are provided, the circuit has a simple structure, the resources are few used, and the cost and the power consumption are reduced.

The invention provides a laser mileage calculation method based on a point cloud multi-view projection drawing, and the method is applied to a mobile robot and an automatic driving platform, and comprises the following steps: 1, carrying out multi-view two-dimensional projection on three-dimensional laser point cloud data to obtain a cylindrical panoramic projection view and an aerial view of an environment, and further segmenting the projection to obtain a multi-view point cloud projection drawing; 2, constructing a multi-view pose estimation network, and extracting environmental geometric structure features on the point cloud projection drawing; 3, calculating a pose loss function based on a pose absolute value and a luminosity reprojection error loss function based on a stable point region; 4, training a neural network model according to the pose loss function, wherein the neural network model can estimate relative pose transformation according to the input point cloud data of the front and back frames.

The invention discloses a video image analysis processing system and device, and belongs to the technical field of video image processing. The system comprises an acquisition unit, a user terminal and an analysis unit, according to the system, sensing data and video data are fused by utilizing two modes of sensing data and video analysis processing, multivariate data fusion is realized, more exercise data are provided for a user, the uncertainty during exercise is made up, and the problems that in the prior art, video data are too single and insufficient in precision, data cannot be quantized, suggestions cannot be provided for a customized scheme are solved; the invention further provides two types of inertial measurement units capable of detecting six-axis data, so that the problem of insufficient precision in the prior art is solved; in addition, the video analysis technology is used for extracting the action time of the resultant force when the baseball is hit, the resultant force of all the external force including the gravity borne by the baseball when the baseball is hit is reversely calculated, force generation data during bat swing are provided and fused with video data, and visual processing is further achieved.

The invention provides a method for measuring the physical yielding point position and the stress level of a component. The method is technically characterized in that thermal image shooting is performed on the component in the static load tension test process, and the physical yielding point position and the stress level of the thin-plate component are accurately measured based on synchronous load and component temperature field data. Relatively accurate physical yielding point position and stress can be acquired, and more accurate mechanical property reference data are provided for welding joint design and performance evaluation. As the position firstly with yielding is the weakest position in a structure, the method can be applied to structural static load test to accurately locate theweak position, overall performances are improved for design, and the problem of insufficient precision of a current macroscopic shielding strength measuring method based on a stress-strain curve is solved. The method is simple to use, easy to implement and excellent in practicability, and original static load tension test is not changed.

The invention discloses a cable drum control method and a cable drum control device. The method comprises the following steps of positioning and calculating the relative distance between a shore cabledrum and a wharf boat power connection box; calculating the actual length of a cable; measuring the tensile force borne by the cable; and judging the stress of the cable, wherein if the tensile forceborne by the cable exceeds a normal range, the winding and unwinding of the cable are controlled according to the deviation between the actual length of the cable and the relative distance. By introducing a cascade control system and using the stress of the cable as a given value for cable length adjustment, the length of the cable can be adjusted according to the stress of the cable, disturbancegenerated by a sudden stress change of the cable is overcome, and the problem of insufficient GPS precision is solved.

The invention discloses a calculation simulation method for soil body large-deformation flow. The method is based on a molecular dynamics (MD) method. In the method, a Newton motion equation is taken as a control equation; a Hertzian type friction formula and a Stokes formula are introduced to describe solid-solid particle interaction and solid-liquid particle interaction respectively; and a solid-liquid coupling soil body large-deformation flow calculation model is built in a process of solving an integral through a Verlet algorithm. Through adoption of the calculation simulation method, a basic dynamics feature and a motion rule of soil body particle complex flow can be presented; the research level of prevention and control of soil body large-deformation flow disasters in China is further perfected and raised; and meanwhile, a certain basis is laid for wide application and popularization of a molecular dynamics calculation method in the soil body large-deformation flow disasters.